The proposed work innovatively addresses serious and long-standing limitations in the evidence supporting the standard relapse model. The important role of the standard model both in pre-clinical research into relapse and in clinical treatments to prevent relapse make it important to strongly test the assumptions of the standard model. This work could result in a strong validation of the standard model and improved animal models for developing medications designed to prevent relapse and craving. Alternatively, this work might indicate that other, less understood, behavioral mechanisms are at play and redirect our efforts towards identifying and understanding the behavioral biology of these mechanisms; and developing new more effective treatments based upon this improved understanding. Finally, this work extends the study of the role of classical conditioning from relapse to recovery, an extension that may aid in the development of new and different approaches to the treatment of problem drinking. While a substantial body of evidence indicates that Ethanol-Paired-Stimuli (EPS) increase ethanol-seeking (in animals) and increase craving (in humans), little evidence is available to show that this increased ethanol- seeking (or craving) is the result of an EPS serving as a CS. Additionally, linking our pre-clinical measure of craving (increased ethanol-seeking) to the actual clinical phenomenon of craving is difficult. Further, this EPS occasioned increase in ethanol-seeking (or craving) has not been shown to result in increased drinking. AIM I of this proposal would examine whether an EPS can increase ethanol-seeking by functioning as a CS, whether this increased ethanol-seeking actually results in increased drinking (and other relapse-like behaviors), and AIM II begins to assess whether an EPS can produce craving-like effects in rats by acting as a CS. This work could result in strong validation of the standard model and improved animal models for developing medications to prevent relapse and craving. Alternatively, this work may indicate that other, less understood, behavioral mechanisms are at work and redirect our efforts towards identifying and under- standing the behavioral biology of these mechanisms; and developing new more effective treatments based upon this improved understanding. Finally, the role of classical conditioning and associative learning in problem drinking almost certainly extends beyond their role in relapse. AIM III begins to address these other roles by examining the role associative learning may play in recovery. By developing animal models of recovery, this project develops new tools for the development of better treatments of problem drinking.